Little is known about data sharing preferences of patients with behavioral health conditions (BHCs) and their surrogates, and even less is known about the choice of patients with Serious Mental Illness (SMI). Our hypothesis is that, instead of wanting the currently used broad consent models, behavioral health patients and surrogates want a greater degree of choice to determine at a granular level which health information (in particular, sensitive information) to share for care and research. The main goals of the proposed research are to address the open challenges of: (1) identifying categories of sensitive health information that are meaningful for behavioral health patients and surrogates, (2) specifying the identified categories of sharing choices in a precise and uniform way by linking them to standard medical terminologies, (3) understanding providers perspective on implementing patients' and surrogates' driven data sharing choices, (4) deploying and testing an electronic informed consent management tool, that we call MyChoice that supports tiered-consent models, (5) helping BCH patients to intelligently exercise granular control by embedding into MyChoice educational material on the type of information contained in each data category and potential favorable and unfavorable consequences of restricting data access for care, and (6) deploying and testing MyChoice and the Trusted Broker in a context of a behavioral health information exchange (HIE). At the end of the study we will recruit 200 adult patients with BHCs (70 with SMI) and 70 surrogates and randomly assign them to: (i) intervention arm: the recruiter is someone involved in the patient's care; (ii) control arm: the recruiter is not part of the facilitis' care team. We will consent both arms with MyChoice, though their choices will have no effect on the actual data sharing. We will investigate if differences occur in sharing options, levels of comprehension and consent between (i) and (ii) and within the population of patients with BHCs, patients with SMI, and surrogates. Also, we will inspect the electronic medical records of the participants and randomly select 50 encounters occurred within 3 months of consenting with MyChoice. For those encounters we will retrospectively use the Trusted Broker to identify the data that providers could have been restricted to access if privacy preferences were respected. We will use those scenarios to analyze clinical, ethical, policy, regulatory and technical implications.